Near-Optimal Beam Selection for Beamspace MmWave Massive MIMO Systems

The recent concept of beamspace MIMO can utilize beam selection to reduce the number of required radio-frequency (RF) chains in mmWave massive MIMO systems without obvious performance loss. However, as the same beam in the beamspace is likely to be selected for different users, conventional beam selection schemes will suffer from serious multiuser interferences, and some RF chains may be wasted since they have no contribution to the sum-rate performance. To solve these problems, in this letter, we propose an interference-aware (IA) beam selection. Specifically, by considering the potential multiuser interferences, the proposed IA beam selection first classifies all users into two user groups, i.e., the interference-users (IUs) and noninterference-users (NIUs). For NIUs, the beams with large power are selected, while for IUs, the appropriate beams are selected by a low-complexity incremental algorithm based on the criterion of sumrate maximization. Simulation results verify that IA beam selection can achieve the near-optimal sum-rate performance and higher energy efficiency than conventional schemes.